Articular cartilage is the gliding surface of diarthrodial joints. The loss of articular cartilage, through trauma or osteoarthritis, is a major cause of pain and disability. Current treatments address the symptoms that result from cartilage loss. Treatments are needed that correct the cartilage loss itself. One approach to accomplishing this is, in theory, to stimulate the cartilage cells with cell-regulatory molecules such as fibroblast growth factor-2 (FGF-2) and insulin-like growth factor-I (IGF-I). A second approach is to employ cartilage structural molecules, such as collagen and aggrecan, with which to build new cartilage. It is not known how these alternative approaches compare with each other, or whether they can be combined. This proposal seeks to answer these questions. The studies will employ cell-based gene therapy in an ex vivo model of cartilage repair to test the following hypotheses. Objective 1: Determine whether gene therapy with fibroblast growth factor-2 (FGF-2) can induce repair-promoting activity without also inducing damage-promoting activity. Hypothesis 1A: The repair-promoting and damage-promoting actions of overexpressed FGF-2 are selectively mediated by specific FGF-2 isoforms and therefore specific FGF-2 isoforms will preferentially augment new cartilage formation. Hypothesis 1B: The mechanism of FGF-2 isoform action involves differential subcellular localization of the isoforms. Objective 2: Determine whether gene therapy using matrix genes augments articular cartilage repair. Hypothesis 2A: Overexpression of type II collagen or of aggrecan increases new cartilage formation. Hypothesis 2B: Overexpression of both type II collagen and aggrecan is superior to overexpression of either alone in generating new cartilage formation. Objective 3: Determine whether gene therapy with growth factor genes and matrix genes in concert is superior to gene therapy employing either class of agent alone in improving articular cartilage repair. Hypothesis 3A: Delivery of selected FGF-2 isoform and IGF-I genes together with collagen and aggrecan matrix genes increases new cartilage formation to a greater degree than either class of agent alone Hypothesis 3B: Human articular chondrocytes respond to gene transfer in a fashion similar to that of bovine articular chondrocytes. Repair will be assessed using histological, biochemical, and molecular biological parameters. The long-term goal of these studies is to select gene therapy agents for cartilage repair that are suitable for clinical application.